Hot exhaust gases from the turbine could be utilized to preheat the compressed air entering the combustion chamber. A recuperator is then installed, and functions as a counter flow heat exchanger, as shown in Figure 2.5.
By preheating the compressed air less heat needs to be produced in the combustion chamber, resulting in less fuel needed and hence the total efficiency is improved. It is important that the pressure ratio is not too high. As an example, by comparing two turbines, they have the same turbine inlet temperatures and different pressure ratios. The pressure of the exhaust gases leaving the turbines is equal. As seen in the T-s diagram in Figure 2.6 the turbine outlet temperature of the turbine operating with a higher pressure ratio is lower than the one with lower pressure ratio. And hence the turbine with the lower pressure ratio is more suitable for recuperating than the one with high pressure ratio. Thus most recuperators are used in small gas turbines such as microturbines, which are characterized by having low pressure ratios.
5
Figure 2.5 Process scheme of a gas turbine with a recuperator
11 The recuperated cycle is illustrated in the T-s diagram in Figure 2.7, it should be noted that it does not take the pressure losses through the recuperator or the combustion chamber into account. As seen in Figure 2.7 heat from the exhaust gases, QR, is used to preheat the compressed air between point 2-3.
For an ideal recuperating cycle the exhaust temperature is equal to the temperature of air leaving the recuperator i.e. T3 is equal to T5 in Figure 2.7. Due to heat losses to the surroundings this does not happen in reality. Assuming the recuperator to be well insulated
s T
ΔTin
4
Qout
1 2
Qin
3 QR
6
5
Figure 2.7 Recuperating cycle s
pl ph2
ph1
TIT T
Figure 2.6 Comparing two turbines
12 and any changes in the kinetic and potential energies to be negligible the maximum heat transfers from the exhaust gases to the air can be expressed as [1]:
(2.13) and
(2.14)
From these equations the effectiveness i.e. how close the recuperator is to an ideal recuperator can be calculated. The effectiveness ε is defined as
(2.15)
Based on Figure 2.7, the total efficiency of the cycle can be expressed as:
(2.16)
As shown in Figure 2.8, the efficiency of the recuperating cycle decreases when the pressure ratio increases. The stippled line symbolizes the efficiency of the simple cycle (as in Figure 2.2) and t is the temperature ratio between T4 and T1. When T2 is larger or equal to T5 the efficiency lines intersect and the recuperator is no longer useful.
Figure 2.8 Efficiency, simple cycle with recuperator [2]
13 2.3 Compressor
In this chapter both centrifugal and axial compressors will be described.
A pressure ratio must be provided in order to produce an expansion through a turbine. The compression of air is therefore a necessary step in the system. As air is a compressible medium, the compressor increases the pressure by reducing the volume of the gas.
The operating range for a compressor is between the surge point and the choke point, as shown Figure 2.9. The surge and choke point depends on the pressure ratio in the compressor and the flow rate.
If the flow rate reduces, it leads to a loss of delivery pressure, and the flow will be reversed.
This is called surge. Choking occurs when the flow increases to the relative speed of sound.
The impeller vanes are not designed to handle this amount of flow, and will cause a loss in delivery pressure. Surge is something that needs to be avoided; mainly because of the forces acting on the compressor when the flow is reversed, may lead to a total destruction. Choke conditions cause a large decrease of the efficiency, but do not lead to destruction of the unit [3].
2.3.1 Centrifugal compressor
A centrifugal compressor consists of a stationary casing containing inlet guide vanes, an inducer, a rotating impeller, a number of fixed diffusers and a scroll. The air comes into the
Mass flow
Figure 2.9 Surge- and chokepoint shown with a constant speed curve
14 compressor through an intake duct and is given a prewhirl by the inlet guide vanes. The air is then sucked into the impeller eye and consisting an inducer where the direction is changed from axial to radial [3]. It is the centrifugal effect causing the air to flow radially outwards along the vanes to the impeller tip [4]. Because of the rotating impeller the air is accelerated and the static pressure increases between the impeller eye and the impeller tip. As the air flow through the diffuser, which is a divergent nozzle, the kinetic energy (velocity) is converted to pressure energy, as seen in Figure 2.10. The compressed air enters the scroll and is discharged.
Figure 2.10 Pressure and velocity through a centrifugal compressor [3]
There will be a small pressure loss due to the friction in the diffuser [2]. The normal practice is to design the compressor so that about half of the pressure rise occurs in the impeller and the rest in the diffuser [4].
The pressure ratio in a centrifugal compressor varies between 3: 1 and 7: 1. This is a relatively small pressure ratio compared to axial compressors which can obtain an overall pressure ratio of 40: 1. Axial compressors need several compressor stages achieving the high pressure ratio, because one stage of the axial compressor only has a pressure ratio between 1,1: 1 and 1,4: 1.
The centrifugal compressors on the other hand, increase the pressure in one stage. Because of this the centrifugal compressor takes less space than the axial compressor, and is a preferable choice in small gas turbines e.g. microturbines.
15 2.3.2 Axial compressor
An axial compressor is usually designed with multiple stages to fulfill required delivery pressure [2]. One stage consists of one rotor and one stator. The air is accelerated by a row of rotating blades and then decelerated by a row of stationary blades. The increase stagnation pressure i.e. the total pressure is accomplished by the rotor. As the air is accelerated in the rotor creating a dynamic pressure, the stator transforms this kinetic energy into an increase of static pressure, by decelerating the air flow as shown in Figure 2.11. The changes in the total conditions for pressure, temperature and enthalpy only occur in the rotating component i.e.
the rotor, where the energy is supplied to the system.
Figure 2.11 Variation of enthalpy, velocity and pressure through an axial compressor [3]
Due to friction there will be a loss in stagnation pressure between the rotor and stator [2].
As seen in Figure 2.11 the length of the blades and the annulus area is decreasing through the length of the compressor. The reason for this is to obtain a constant axial velocity, as the reduction of flow area compensate for the increase of density of the air as it is compressed [3].
As the axial compressor can have an overall pressure ratio of 40: 1, it dominates the field for industries needing large power [2].
2.4 Combustion chamber
In the combustion chamber the heated air is mixed with fuel e.g. natural gas or liquid petroleum distillates. At start up of the gas turbine the mix of fuel and air needs to be ignited by electric spark to initiate the combustion process. Thereafter the flame must be self-sustaining [2].
16 2.4.1 Combustor design
The design of the combustion chamber depends on whether the gas turbine is on an aircraft or on a ground-based system.
The can-type combustor consists of a number of separate chambers spaced around the shaft connecting the compressor and turbine, as shown in Figure 2.12. The air from the compressor is split into separate streams, each supplying a separate chamber. This arrangement is suitable for engines containing centrifugal compressors where the flow is divided into separate streams in the diffuser [2]. The advantage of can-type combustors is that they are easy to test and maintain, because it can be done on one single can rather than the whole combustion system. Can-type combustors were widely used on early gas turbines. Due to their high weight and relative large pressure drop, most modern gas turbines do not use can combustors today.
Figure 2.12 Can-type combustor [4]
Recent designs make use of cannular combustion chambers. As with the can-type combustor the cannular design also have separate combustion cans. It has individual flame tubes spaced around an annular casing [2], as illustrated in figure 2.13. Because of the annulus casing each can does not have to serve as a single pressure vessel, which results in a smaller pressure drop than the can-type design. The cannular combustion chamber is the preferred combustor type in most gas turbines. The reason for this is the relative low pressure drop, good temperature distribution and they are easy to maintain.
17
Figure 2.13 Cannular combustor [4]
The most compact combustion design is the annular combustor. This type of combustion consists of a single flame tube, which is contained in an inner and outer casing [4], as shown in Figure 2.14. Because of the compact design the combustor requires less space and saves a considerable amount of weight. A major advantage of the annular design is the wall area surrounding the system is much less than the systems described above, this results in less amount of cooling air and a higher turbine inlet temperature. Because of its low weight and low frontal area it is a preferred choice in aircraft engines.
Figure 2.14 Annular combustor [4]
2.4.2 Methods for reducing emissions
When burning fuel it is desirable to achieve complete combustion to prevent dissociation of carbon monoxide (CO) and unburned hydrocarbons (UHC). CO is a very toxic pollutant that
18 must be controlled to very low levels. The amount of CO and UHC produced is increased if the flame temperature is low. NOx on the other hand is produced when the combustion temperature is high as illustrated in Figure 2.15. However, if the residence time of the fluid in the combustor is increased the formation of UHC and CO will be decreased, due to achievement of complete combustion. Increasing the residence time implies an increase in combustor cross-sectional area or volume [2].
Due to high combustion temperature NOx is produced. In recent years the legislations of NOx emissions has been stricter, and has led to significant changes in the combustor design. The standard level of NOx set by Environmental Protection Agency (EPA) is 75 ppmvd (parts per million by volume of dry exhaust gas). Basically there are three major methods of minimizing emissions: water or steam injection into the combustor, selective catalytic reduction and dry low NOx [2].
The purpose of water/steam injection is to reduce the flame temperature. It is important that the water is de-mineralized to prevent corrosive deposits in the turbine. In the first installations there was injected half as much water as fuel. This resulted in a 40 % reduction in NOx. The amounts of water required today are substantial. Thus this type of installation is not a preferable choice in locations where water recourses are scarce and expensive. Also in locations where the ambient temperature is below freezing point there must be used other types of installations to reduce the NOx emissions. The power output will have a small increase as a result of a higher mass flow through the turbine, but the there will be a decrease in thermal efficiency, because the TIT will be lowered. It has also been discovered that while continuing to decrease NOx, there will be an increase in both CO and UHC. The reason for this is when the flame temperature is decreased there will be an increase of CO and UHC as
NOx
Temperature CO & UHC
Emissions
Figure 2.15 Formation of CO, UHC and NOx as a function of tempeature
19 illustrated in Figure 2.15. Steam injection operates at the same principle as water injection and is applied in systems producing steam e.g. cogeneration plant and combined cycles [2].
Selective catalytic reduction (SCR) has been used in applications where the specified limits of NOx are extremely low. This unit is installed after the turbine cleaning up the exhaust gases.
A catalyst is used together with injection of controlled amounts of ammonia (NH3) and results in a conversion of NOx to N2 and H2O. This conversion can only occur in a limited temperature range (285 – 400 °C) and the system is installed in the heat recovery steam generator. Thus it can only be used in applications containing a waste heat recovery system i.e. combined cycle systems. The difficulties with SCR are controlling the NH3 when dealing with variable loads and the handling and storage of the noxious fluid. The cost of using SCR system needs also to be taken into account [2].
The term dry low NOx derives from the fact that there is no water involved when reducing the NOx emissions. With this design the air and most of the fuel is premixed before entering the combustion chamber. The fuel/air ratio is then lean which reduces the flame temperature and hence the NOx emissions are decreased [2]. The injector in the dry low NOx technology has two fuel circuits, one (approximately 97 %) premixing the fuel with air and one injecting the rest of the fuel directly into the combustion chamber. The reason for this is when the gas turbine operates on part load the mixture of fuel/air can be too lean to burn, and a flame out can occur. A small portion of the fuel is burned richer in order to have a stable flame. A swirler is used to create the required flow conditions in the combustion chamber to stabilize the flame [3].
2.4.3 Cooling systems
The combustor outlet temperature can be as high as 1850 K [2], this sets some requirements to the materials and cooling. The liner is the inner part of the casing which is exposed to the high temperatures radiated from the flame and combustion. To improve the life of the liner, it is necessary to lower the temperature of the liner and use a material that is resistant to thermal stress and fatigue. The air film cooling method reduces the temperature of the surface both inside and outside the liner. A metal ring is fastened inside the liner making an annular clearance. Some of the compressed air is used as coolant between the surfaces of the liner. In systems containing a combined cycle steam could be used as coolant instead of air.
The material of the liner being used is Nimonic 75, which is an 80-20 nickel-chromium alloy stiffened with a small amount of titanium carbide. This material has an excellent oxidation and corrosion resistance at elevated temperatures, a good resistance against fatigue and
20 reasonable creep strength. As the firing temperature has increased in the newer gas-turbine models, HA-188 has been employed. HA-188 is a chromium-nickel alloy which has improved the creep rupture strength. Many of today’s combustors also have thermal barrier coatings.
radial flow and axial flow. In this chapter both types will be described.
2.5.1 Radial flow turbine
Basically the radial flow turbine is a centrifugal compressor with reversed flow and opposite rotation [3]. The gas flows with a high tangential velocity directed inwards, and leaves the rotor with as small whirl velocity as possible near the axis of rotation [2]. The appearance of the radial flow turbine is very similar to the centrifugal compressor, but instead of diffuser vanes, the radial flow turbine has a ring of nozzle vanes [2]. The nozzles transform the dynamic pressure into kinetic energy. To increase the efficiency of the turbine, the turbine outlet is connected to a diffuser [3]. The outlet diffuser converts the high absolute velocity into static pressure.
The work produced by a single stage radial turbine is equivalent to the work produced by two or more stages in the axial turbine. This phenomenon occurs because a radial flow turbine usually has a higher tip speed than an axial flow turbine. As the power output is a function of the square of the tip speed, the power produced by a radial flow turbine is greater than a single stage axial turbine [3].
In compact designs the gas turbine often consists of a centrifugal compressor mounted back-to-back with a radial flow turbine. The rotor then becomes short and rigid and hence more efficient. Radial flow turbines have also been widely used as a turboexpander in the cryogenic industry and in turbochargers for reciprocating engines [2]. In these systems the mass flow rate and pressure ratio is low. The radial flow turbine handles low flow rates more efficiently than axial flow.
In order for radial turbines to be cost effective they are installed without cooling system. A cooling system on this type of turbine would have to be very complex, due to the turbine wheel. This sets some limitations to the turbine inlet temperatures and material being used in
21 order for the turbine to sustain. Nickel alloys such as Inconel is often being used as turbine material. These materials can handle TITs up to 1000 °C.
Because of the cheap, compact and robust design, the radial flow turbine is a preferable choice in microturbines.
2.5.2 Axial flow turbine
The vast majority of gas turbines employ the axial flow turbine. This is because the axial turbine is most efficient in most operational ranges [3].
The axial flow turbine can consist of one or more stages. As with the axial compressor, one stage of the axial flow turbine consists of one stator and one rotor. The stator consists of a row of nozzle blades, which decrease the pressure and temperature and increases the velocity. The kinetic energy from the stator is converted to mechanical energy through the rotor blades.
There are two types of axial flow turbines: The impulse type and the reaction type. In the impulse type turbine the entire enthalpy drops in the nozzle. This causes the flow to have a high velocity entering the rotor. The reaction turbine divides the enthalpy drop in the nozzle and the rotor. The impulse stage produces twice the output of a comparable 50 % reaction stage, but the efficiency is less than that of a 50 % reaction stage. The cost of a reaction turbine for the same amount of work as an impulse turbine is much higher this is because the reaction turbine requires more stages. A multistage axial turbine usually consists of impulse type turbines in the first few stages and reaction type turbines in the later stages. Building the turbine this way, the pressure drop in the first stages is maximized and a good efficiency is obtained [3].
2.5.2.1 Turbine blade cooling
As axial turbines are used in large power plants that can produce hundreds of MW, they operate with high turbine inlet temperatures (TIT). Referring to equation 2.12 in chapter 2.1.3 it is desirable to have as high TIT as possible. The challenge for engineers today is to find materials and turbine blade cooling techniques that can handle these high temperatures. TIT could be higher than 1400 °C and the turbine blade alloys can only handle a temperature at approximately 870 °C. Advanced air cooling is needed to prevent destruction of the turbine components. There are five basic air-cooling schemes: Convection cooling, impingement
As axial turbines are used in large power plants that can produce hundreds of MW, they operate with high turbine inlet temperatures (TIT). Referring to equation 2.12 in chapter 2.1.3 it is desirable to have as high TIT as possible. The challenge for engineers today is to find materials and turbine blade cooling techniques that can handle these high temperatures. TIT could be higher than 1400 °C and the turbine blade alloys can only handle a temperature at approximately 870 °C. Advanced air cooling is needed to prevent destruction of the turbine components. There are five basic air-cooling schemes: Convection cooling, impingement